The present disclosure relates to improvements in the systems and method used to recommend a course of action to a user.
Recommender systems and algorithms such as, e.g., collaborative filtering and content based recommenders are widely used in many domains including the recommendation of books and movies, courses of study, jobs to apply for, skills to learn, fitness activities, etc.
Often recommendations output by a recommender system are generally shown to a user with little explanation or motivation for why the action is recommended for that specific user.
Some example attempts to improve on this lack of information include providing an explanation or indication for why a user may want to take a certain action. For example, the recommendation may include a comparison of the action to a similar action previously taken by a user, such as recommending a movie because the user previously watched another similar movie. As the information available to a user increases, such explanations have become an increasingly important way for the user to accept and understand the recommendation.